It is tempting to imagine that the nave of Reims Cathedral was originally planned to have a cross-section very similar to that of Soissons Cathedral, with simple blocky buttress uprights whose inner margins would rise directly above the inner margins of the wall shafts below, as shown at left. The lower set of flying buttresses would intersect the clerestory wall right at the springers of the main vault. The choir, seen at right, would have had a similar system, but extended to span the double aisles. The heights of the inner and outer uprights in the choir are used here to set the upper and lower surfaces of the nave uprights, respectively. The overall height of the main vessel is here shown as exactly twice the height of the side aisles, as would be seen in the Reims elevations drawn by Villard de Honnecourt.
At some point probably in the 1220s, the hypothesized original buttress design was revised in favor of the pointier design recorded by Villard de Honnecourt. The proportions of Villard’s drawing are too narrow to fit the building, as many scholars have noted, but if the drawing is stretched laterally, its relative proportions actually fit the choir fairly well, at least when its overall height is set at twice the aisle height, as shown here. It is clear, in any case, that pinnacles were foreseen atop both the main and intermediate buttress uprights in the design he records, even if the pinnacle format differs from the variants actually built.
In the original design, and still during the time of Villard’s visit to the Reims workshop, the main vessel was meant to be twice as tall as the aisles, which means that it would have fit perfectly within an octagon whose bottom facet coincided with the span between the main arcade axes at ground level. It is tempting to imagine that the height to the top ridge of the roof was originally meant to be √2 times greater than the height of the octagon, a dimension that can be found by unfolding the diagonal of the square circumscribed around it.
At some point probably soon after Villard’s visit, and before the completion of the transept facades, the pinnacle design was modified to the current spirelike format, and the design of the buttress uprights was revised to their current slender format, as seen at right. Because these uprights are much narrower than those of the hypothesized original design seen at left, their inner margins are displaced outward with respect to the shaft bundles below. As a consequence, the radius of the flying buttress arches grows larger. The flyers and pinnacles seen at left are exactly as they are today, but they have been positioned lower, as the following slide will show.
Here the current state of the cathedral is seen at left, contrasted with the hypothesized penultimate version of the nave, at right. In the current design, the main vault has been made sharper than before, so that it reaches a greater total height even though the capitals and springers are located as in the previous schemes. This means that the clerestory wall grows taller, and the flying buttress pair moves upward as a consequence. As a result of this shift, even the bottom of the lower flyer now abuts the wall significantly above the vault springer, a position unseen at similar buildings (e.g. Chartres, Soissons, Amiens, Cologne, etc.). This shift also explains the presence of the rather awkward blocks elevating the flyer springers above the top of the aisle roof, which were not necessary in the previous scheme seen at right.
The drawing of the buttresses by Villard de Honnecourt comes from folio 32v of Bibliothèque nationale de France. Département des Manuscrits. Français 19093.
The drawn section of the choir comes from Thomas H. King, The Studybook
of Medieval Architecture and Art (London: Bell and Daldy, 1858)].
The section of the nave is based on a laser scan made in 2018 by Pierre Hallot of the University of Liege, working in collaboration with Robert Bork, Adam Skibbe, Michelle Wienhold, Drew Hutchinson, and Rebecca Smith of the University of Iowa.